Precision medicine (PM) represents a paradigm shift in health care – moving from generalized treatment toward individualized care informed by each person’s genetic, environmental, and lifestyle profile. The advent of artificial intelligence (AI) and deep learning (DL) has accelerated this transformation by enabling the integration of vast and heterogeneous biomedical data. This review explores how DL has advanced the analytical power of genomics, radiology, and biomedical imaging, forming the technological core of modern PM. In genomics, AI facilitates variant classification, gene expression modeling, and multiomics data fusion for disease risk prediction. In radiology and medical imaging, convolutional and transformer-based architecture enhance lesion detection, image reconstruction, and radiogenomic mapping. Emerging applications in ophthalmic imaging and digital pathology further demonstrate AI’s role in early diagnosis and personalized therapy planning. Despite remarkable progress, challenges remain in ensuring data interoperability, algorithmic transparency, and ethical governance. The review discusses solutions such as federated learning, explainable AI, and privacy-preserving computation that are essential for trustworthy implementation. Looking ahead, the integration of these technologies aligns with the Society 5.0 vision, fostering a human-centered, sustainable, and equitable healthcare ecosystem where intelligent PM bridges technology, ethics, and global well-being.
Athletic training and menstruation both contribute to iron depletion and anemia in female athletes. We hypothesized that oral iron supplementation during the menstrual period would increase the hemoglobin (Hb) level and reduce fatigue in female athletes. Twenty-two female athletes were randomly assigned to either the control (CON) or experimental (EXP) groups. Participants in the EXP group received one capsule of ferrous bisglycinate chelate (equivalent to 20 mg of elemental iron) daily during their menstrual periods in the second, third, and fourth menstrual cycles (M2-M4), whereas those in the CON group took a placebo. Blood samples from each subject were collected after each menstrual period, and Hb, hematocrit, and blood cell counts were analyzed. The results indicated that iron supplementation significantly increased the levels of Hb, hematocrit, and red blood cell (RBC) count from the first menstrual bleeding period (M1) to the fifth menstrual bleeding period (M5). The immune-related cell populations and rating of perceived exertion fatigue indicators showed no significant differences from M1 to M5. In conclusion, daily supplementation with ferrous bisglycinate chelate during the menstrual period significantly increased Hb levels, hematocrit levels, and RBC counts in female athletes.
The properties and dynamics of small biomolecule pools of a cell or an organ have a significant effect on the metabolism of the related biomacromolecules and the ongoing life process. Here, we investigated the properties of the nucleotide pool and differentially expressed transcriptome in the liver of CCl4-induced (CI) mice for fibrosis and explored the relationship between the two. We used high-performance liquid chromatography to detect the contents of nucleotide pool in control and treatment groups, while the changed gene expression data in fibrosis progression were collected from BioGPS, and the change ratios of differentially expressed genes were mined out from different literatures about CI mice for fibrosis. All data, expressed as mean +/- standard deviation, were analyzed with the SPSS. The main results were as follows: (1) the performance of nucleotide pool of CI livers was quite different from that of control livers, i.e., the contents of the total nucleotides and nucleoside mono phosphates (NMPs, the dominant components of the pool), and the property of nucleotide pool was associated with the physiological state as well as its correlation coefficient was 0.904 (P < 0.05) and (2) the property of nucleotide pool was associated with the differential transcriptome during fibrosis, the variable quantity of four kinds of nucleotides (NMP) within the nucleotide pool and the counterparts of differentially transcriptome of CI mice have the significantly positive correlation with the correlation coefficient as 0.927 (P < 0.05). The important connection of the nucleotide pool with the differentially transcriptome and the liver fibrosis was revealed for the first time. The new mechanism of exogenous nucleotides that had an effect on fibrosis was revealed.
Systemic chemotherapeutic drugs such as paclitaxel (PAC) and lenvatinib (LEV) are widely utilized in the treatment of solid tumors. Although their pharmacological detection methods have been well documented, the growing interest in lipid-based nanoparticle (LNP) delivery systems necessitates the advancement of robust methods for in vivo quantification. In this study, we developed and validated a method of liquid chromatography coupled with electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) for the quantification of PAC and LEV encapsulated in peptide-guided LNP (PGsLNP-PAC and PGsLNP-LEV). Initial method development was performed in vitro, followed by in vivo analysis using mice administered either the native drugs or the drugs encapsulated in nanoparticles. The quantification of PAC and LEV was within the linear ranges of 2-500 ng/mL and 0.08-20 ng/mL, with correlation coefficients (R2) of 0.9986 and 0.9998, respectively. The detection was performed using electrospray ionization positive ion mode and quantified by multiple reaction monitoring (MRM), which showed strong signal intensities for the parent ion (m/z) transition pairs 876 -> 308 and 427 -> 370 for PAC and LEV, respectively. The stable-isotope-labeled internal standards 6 alpha-OH-PAC-d5 and LEV-d4 were used with transition pairs 897 -> 313 and 431 -> 370 in MRM mode to correct the errors. The inter- and intra-day precision of low quality control (QC) to high QC for PAC and LEV ranged from 2.0%-18.1% to 0.9%-6.9%, respectively. The method was successfully applied by administering 200 nM of drugs for the quantification of PAC and LEV in tissue lysates. This ultra-sensitive approach enabled the accurate detection of trace-level concentrations of PAC (0.1810 mg +/- 0.27/mL) and LEV (0.0066 mg +/- 0.030/mL) in PGsLNP-PAC and PGsLNP-LEV. Furthermore, the method was validated by measuring the concentrations of native PAC in intravenously-administered C57BL/6 mice tissues. The reticular endothelial system, i.e. lung, spleen, and liver, showed a higher concentration of 396.1, 67.4, and 48.2 ng/g, respectively. We demonstrated a sensitive and robust LC-ESI-MS/MS method that can be suitable for pharmacokinetic evaluation and nanoparticle-based drug delivery assessment in cancer therapy.
The 17 alpha-ethinylestradiol (EE2) is a derivative of estradiol, which has been used to be the oral contraceptive pills. EE2 is an environmental hormone and influences the endocrine system in animals. In the present study, we explored the mechanism of EE2-modulatory corticosterone production in vitro and invivo. EE2-pretreatment not only decreased the corticosterone production from rat zona fasciculata reticularis (ZFR) cells but also led to a reduction of the stimulatory effect of adrenocorticotropic hormone (ACTH) on the maximum level of plasma corticosterone in rats. Administration of EE2 (10-40 mu g/ml/kg) in vivo dose dependently decreased the basal and 8-bromo-cyclic AMP (8-Br-cAMP)-induced production of corticosterone from rat ZFR cells invitro. The in vitro corticosterone production in response to ACTH was also attenuated after in vivo administration of EE2. The steroidogenic acute regulatory (StAR) protein, a protein transports cholesterol from outer to inner mitochondria membrane, and P450scc, a steroidogenesis-limiting enzyme, were downregulated after treatment with EE2 invivo. Other results revealed that EE2 significantly increased the basal levels of pregnenolone but decreased the pregnenolone levelsin response to 25-OH-cholesterol (a biologically active derivative of cholesterol; 10-6 M) and trilostane (an inhibitor of 3 beta-hydroxysteroid dehydrogenase). Moreover, EE2 affected the negative feedback regulation of hypothalamus-pituitary-adrenal (HPA) axis through downregulationof glucocorticoid receptor protein level in the anterior pituitary gland but not in rat medial basal hypothalamus. These results suggest that EE2 may affect the HPA axis to inhibit P450scc activity, basal and ACTH-induced corticosterone production from rat ZFR cells through a cAMP-dependent pathway
Emerging evidence indicates a critical role of an imbalanced microbiota ecosystem in the development of human diseases. The presence of decreased beneficial microbes and increased harmful bacteria is widely accepted as one of the etiologies of intestinal and extraintestinal disorders. Pathobionts (commensal-derived opportunistic pathogens) with colitogenic and tumorigenic capabilities are now recognized as contributors to the pathogenesis of inflammatory bowel diseases (IBDs) and colitis-associated colorectal cancers (CRC). Aside from classical wet-lab techniques and traditional scientific paper-mining approaches, the innovation of artificial intelligence (AI)-powered biotechnology devices and tools, such as machine learning, deep learning, and natural language processing, is a valuable asset for modern researchers to enhance data interpretation and develop precision medical strategies. This review discusses potential AI-assisted approaches in gastrointestinal (GI) pathophysiological research, with a focus on investigating gut dysbiosis. The advantages and shortcomings of AI-generated data and research strategies will be assessed in the context of dysbiosis-driven IBD and CRC. Studies that apply methods leveraging advanced computational techniques and AI-assisted research methodologies would help better understand the complex interactions within the microbiome and their interactions with hosts. A multidisciplinary approach to GI pathophysiological research that integrates AI tools can pave the way for innovative interventions targeting bacteria, with a focus on colitis and cancer.
Artificial intelligence (AI) is transforming physiological sciences by connecting molecular discovery, biomarker identification, real-time physiological signal analysis, and clinical decision support. This review highlights four major domains where AI has achieved transformative progress, bridging this continuum. In structure-guided and generative therapeutics, models such as AlphaFold and deep generative design frameworks enable atomic-level prediction of protein structures, ligand docking, and automated synthesis planning. In biomarker discovery, deep learning applied to proteomic, metabolomic, glycomic, and lipidomic data expands molecular annotation and reveals disease-associated signatures beyond conventional detection limits. AI-driven systems for wearable and implantable monitoring now translate continuous physiological signals into real-time cardiovascular, respiratory, and metabolic insights, advancing precision health assessment. Finally, clinical decision-support frameworks and patient digital twins integrate predictive modeling and mechanistic constraints with evidence reasoning to enhance medical interpretation. Collectively, these advances illustrate how AI has evolved from analytical computation to mechanistic physiological understanding, inaugurating an era of “Intelligent Physiology” that links molecular mechanisms with individualized clinical care.
Through controlling the flow of molecules between the circulation and the brain, the blood–brain barrier (BBB) maintains homeostasis in the central nervous system (CNS). Crucial for sealing endothelial cell connections and maintaining BBB integrity are tight junctions (TJs), which are composed of transmembrane proteins (occludin, claudins, and junctional adhesion molecules). This review critically examines the molecular structure and regulatory processes of TJs within the BBB, emphasizing their dynamic modulation in neurological disorders such as neurodegenerative, neuroinflammatory, autoimmune, cerebrovascular, neuropsychiatric, neurodevelopmental, neuro-oncological, and neurovascular disorders. Emphasis is placed on how TJ disruption contributes to barrier breakdown and disease progression, with insights into key signaling pathways and regulatory networks, including Wnt/β-catenin, CmPn/CmP steroid signaling networks, phosphorylation cascades, matrix metalloproteinase activities, microRNA regulation, and the gut–brain axis. Recent advances in elucidating TJ protein interactions and their regulatory modulators have identified promising therapeutic targets for restoring BBB function and attenuating neuroinflammatory processes. Furthermore, innovative approaches for precisely regulating TJs to enhance drug transport across the BBB are thoroughly examined. By integrating multi-omic studies and the application of artificial intelligence, future directions aim to unravel the complex TJ regulatory network for precision medicine approaches in CNS disorders. Overall, this literature review, which integrates current evidence and novel data from reputable databases, highlights the crucial roles of TJs concerning CNS health and disease. It is expected to promote further investigation of TJ physiology as a basis for innovative diagnostics and therapies to enhance neurological care.
Programmed cell death, or apoptosis, is a critical mechanism contributing to the development of sex differences in cell number within specific regions of the vertebrate brain. While apoptosis has been previously documented in the developing mouse hippocampus and cerebral cortex, detailed analyses comparing both sexes have been limited. In this study, we performed immunohistochemistry for activated caspase-3, a marker of apoptosis, to quantify the distribution of apoptotic cells across subregions of the mouse hippocampus and surrounding cortex at postnatal (PN) days 0 and 7. In addition, we assessed the expression of preselected apoptosis-related genes in the developing hippocampus and cortex using Qiagen's polymerase chain reaction (PCR) arrays and reverse transcription quantitative PCR. Our findings revealed that, during early development, male mice exhibited a higher number and/or density of activated caspase-3 cells in the stratum pyramidale of the cornu ammonis (CA) 1, the stratum oriens of the CA3, and the polymorphic layer of the dentate gyrus. We also observed widespread, age-dependent changes in apoptosis across various hippocampal and cortical subregions. Furthermore, while no sex differences were detected in gene expression, we observed an age-dependent decrease in the expression of Akt1 and Bok mRNA in the developing hippocampus and cortex. Together, our findings reveal region-specific patterns of apoptosis in the developing mouse hippocampus and cortex, influenced by age and sex, and identify apoptotic genes that may play a critical role in the neural development of these brain regions through transcriptional regulation.
Most of the available pharmacological strategies of pain relief use drugs that have monomodal mechanisms of analgesia. However, pain treatment requires combinations of drugs with nonredundant mechanisms of action to improve pain management and reduce the adverse effects produced by high doses of individual drugs. This study aimed to investigate, through isobolographic analysis, whether statins can modify the antinociceptive effect produced by diclofenac. The above was measured by the formalin test in rats, a murine animal model of acute inflammatory pain. Animals were administered with diclofenac (0.3, 1, 3 and 10 mg/kg; p.o.), atorvastatin (1, 3, 10 and 30 mg/kg; p.o.), pravastatin (1, 3, 10 and 30 mg/kg; p.o.), rosuvastatin (1, 3, 10 and 30 mg/kg; p.o.), or calculated combinations of diclofenac with the different statins. The effective doses 30 (ED30) were calculated for diclofenac and statins, and subsequently, the isobolographic analysis of each one of the combinations was performed, thus determining their respective ED30. The experimental ED30 values of antinociception for diclofenac coadministered with atorvastatin, pravastatin, or rosuvastatin were 0.27 +/- 0.01, 1.30 +/- 0.08, and 0.59 +/- 0.08, respectively, while the corresponding theoretical ED30 values of antinociception were 5.08 +/- 0.33, 5.30 +/- 0.33, and 6.64 +/- 0.3, respectively, being statistically higher (P < 0.05) theoretical ED30 values than their corresponding experimental ED30 values. The above suggests that statins enhance the antinociceptive effect of diclofenac. These pharmacological combinations may help treat pain.
ABSTRACT:Metabolic dysfunction-associated steatotic liver disease (MASLD) inhibits hepatitis B virus (HBV) activity while simultaneously exacerbating liver fibrosis/cirrhosis (LF/LC) in chronic HBV (CHB)-infected patients. To date, no model is available to investigate this discrepancy. The established HBV-MASLD-LF/LC mouse model in this report mimics the promotion of LF/LC and suppression of HBV levels by MASLD. Very low-density lipoprotein (VLDL)-loading triacylglyceride (TAG) was positively correlated with the HBV titer and LC/LF in both HBV+ patients and the HBV-MASLD-LF/LC mouse model. TAG treatment could upregulate the HBV titer and fibrotic markers in vitro, thus demonstrating a causal relationship. Hepatocyte-specific VLDL receptor knockout (VRKO) reduced the HBV titer but promoted LF/LC in the HBV-MASLD-LF/LC mouse model. VLDL-TAG was reduced in VRKO HBV-MASLD-LF/LC mice compared to wild-type mice. The VLDL level and VLDL-loading are considered LF/LC risk factors in hepatitis B e antigen-negative CHB patients (considered at low risk of developing LF/LC). In conclusion, this report explains the discrepancies in HBV, MASLD, and LF/LC at the physiological level. VLDL-TAG is considered a novel risk factor of HBV reactivation and deserves further study.
Evidence suggests that changes in gene expression play an important role in the development and progression of Alzheimer's disease (AD). Although many genes have already been identified, there are other genes potentially involved in this process but have not been identified. To further investigate the differences in gene expression profiles between amyloid precursor protein/presenilin 1 (APP/PS1) transgenic mice and wild-type (WT) mice, we have employed RNA sequencing to analyze the transcriptome and identified candidate genes involved in the pathogenesis of AD. Because melatonin was shown to alleviate the pathology of AD and rescue the cognitive impairment in animal models of AD, we have also included APP/PS1 mice that received melatonin treatment for comparison of gene expression profiles. Our results reveal that the E2F transcription factor 8 (E2f8) gene is differentially expressed between APP/PS1 mice and WT mice, and E2F8 contributes to the pathogenesis of AD. This is probably due to E2F8-mediated suppression of the expression of matrix metalloproteinase-9 and B-cell lymphoma 2 that results in amyloid-beta accumulation and neuronal apoptosis. [GRAPHICS]
Diabetic retinopathy remains a leading cause of vision impairment worldwide, with early stages characterized by endothelial dysfunction and later stages marked by pathological neovascularization. Although intravitreal antivascular endothelial growth factor (VEGF) therapies are effective in advanced stages, they show limited efficacy in correcting early vascular dysfunction and macular edema. This study explores the therapeutic potential of GYT-088, a metabolite derived from medicinal fungi, in restoring retinal vascular function at noncytotoxic concentrations. In a type 2 diabetic rhesus monkey model that closely resembles human disease, systemic GYT-088 (6 mg/kg) improved macular edema in two of three animals without altering glucose metabolism. Retinal vessel tortuosity and perivascular changes were also reduced. In a murine model of laser-induced neovascularization, intravitreal GYT-088 suppressed pathological angiogenesis in a dose- and time-dependent manner, showing efficacy comparable to the anti-VEGF agent Eylea. Optical coherence tomography imaging further confirmed preservation of retinal layer structure and reduction of hyperreflective foci. In vitro studies revealed that GYT-088 upregulated endothelial nitric oxide (NO) synthase and claudin-5, while suppressing caveolin-1 (CAV1) expression - leading to increase NO production and reinforced endothelial barrier function. Mechanistic experiments confirmed CAV1 as a negative regulator of GYT-088-mediated vascular protection. Together, these findings support GYT-088 as a promising therapeutic candidate for improving nonproliferative and proliferative diabetic retinopathies.
Brucellosis, a globally significant zoonosis, demands rapid, accurate diagnostics to optimize therapeutic intervention and containment. Traditional detection methods face critical limitations, including poor discrimination between acute and past infections and lengthy testing times. In recent years, detection technologies based on surface-enhanced Raman scattering (SERs) have emerged as a focal point in the field of Brucella diagnosis, offering high sensitivity, real-time readouts, and portability. This review highlights pioneering SERS applications, particularly its synergy with lateral flow immunochromatography, detailing their mechanistic basis, diagnostic metrics, and clinical translational prospects. In addition, we further discuss the advantages and current limitations of SERS technology in disease staging, rapid screening, and deployment in resource-limited settings, drawing on the latest research findings to provide theoretical support and practical guidance for the advancement of Brucella detection methods.
